Complexity Avoidance using Biological Resemblance of Modular Multivariable Structure

Abstract

Combination of Fuzzy logic and Genetic Algorithm is becoming popular among researchers in the field. GFT (Genetic Fuzzimetric Technique) is of no exception which merges Fuzzy logic with genetic algorithm to achieve the optimization of the decision making process under uncertainty. Multivariable structure of any fuzzy rule based system would add complexity when modeling the behavior of the system. The objective of many researchers in the field is to minimize this complexity and enhance the accuracy. The proposed technique is designed to be a modular approach by resembling biological structure where it avoids the complexity instead of minimizing it while attaining acceptable accuracy of the system. After reviewing the mechanism of GFT, a sample application to measure the CRM performance analysis was used as a vehicle to demonstrate the technique. A generic tool termed as Fuzzy Inference Engine (FIE) was built to demonstrate the multivariable modular approach used to implement the CRM performance measurement.